Commit Graph

27229 Commits

Author SHA1 Message Date
krzychb
2ec35f60ca RMT API docs update, new example 2017-11-21 22:34:23 +01:00
Ivan Grokhotkov
c03b068931 spiffs: make OBJ_META_LEN configurable, make mtime support optional
- SPIFFS_OBJ_META_LEN can be set in sdkconfig
- mtime support can be enabled in sdkconfig, if META_LENGTH is sufficient
- add test for mtime updates
2017-11-22 00:17:11 +08:00
luc lebosse
58f046715e Unify the time file creation for SPIFFS and SD 2017-11-21 22:51:36 +08:00
Piyush Shah
4f33339c1d test_ringbuf: Add tests for arbitrary length ring buffer
This will test the ring buffer for buffer length that is not
a multiple of 4

Signed-off-by: Piyush Shah <piyush@espressif.com>
2017-11-21 17:18:54 +05:30
Piyush Shah
7dd9c4f57f ringbuf: Fixes to header files for better understanding
1. Usage of this module required applications to include additional
files. What files to include is not very intuitive. Instead, it is
better for the header file itself to include other files as required.
2. Even though it may seem logical, it is better to explicitly mention
that an item needs to be "Returned" after a Receive

Signed-off-by: Piyush Shah <piyush@espressif.com>
2017-11-21 17:16:04 +05:30
Piyush Shah
62f44e45df ringbuf: Bugfixes for managing arbitrary sizes of ring buffer
It was observed that if the ring buffer size provided by application
is not a multiple of 4, some checks were failing (as read_ptr and write_ptr
could shoot beyond the ring buffer boundary), thereby causing asserts.
Simply wrapping around the pointers for such cases fixes the issue.

Moreover, because of the logic for maintaining 4-byte boundary,
it was also possible that a wrap-around occurred for some data,
even when the actual size remaining was sufficient for it.
Eg. Buffer available: 34, data size: 34, 4-byte aligned size: 36
Since the logic compares against 36, it writes 34 bytes and does a
wraparound. But since all 34 bytes are already written, there is
nothing to write after wrapping. It is better to just re-set the
write pointer to the dtart of ring buffer in such cases.

Tested send/receive for various arbitrary sizes of data and for
arbitrary sizes of ring buffer.

Alternative Solutions:
1) Allow only sizes which are multiples of 4, and return error otherwise.
Appropriate code and documentation change would be required
2) Allow arbitrary sizes, but internally add upto 3 bytes to make
the total size a multiple of 4

Signed-off-by: Piyush Shah <piyush@espressif.com>
2017-11-21 17:15:53 +05:30
Ivan Grokhotkov
0f503dee50 toolchain: update to 1.22.0-75-gbaf03c2
- enable condition variable support in gthread
- add provisions to allow emergency exception pool size to be configured
2017-11-21 19:30:52 +08:00
Ivan Grokhotkov
6e1453e864 Merge branch 'feature/bot' into 'master'
CI: support customize test with @bot

See merge request !1502
2017-11-21 18:43:56 +08:00
Amey Inamdar
78ed138453 cxx_exceptions: Add missing stubs for cxx exceptions with exceptions
disabled

The misssing stubs get pulled in from the toolchain's libstdc++ also
causing to pull in defined stubs. This results in redefinition of
symbols. Fixing it by simply adding the missing stubs when exceptions
are disabled.

Signed-off-by: Amey Inamdar <amey.inamdar@gmail.com>
2017-11-21 12:41:49 +05:30
Ivan Grokhotkov
46673bb6f1 Merge branch 'bugfix/rtc_apll_disable' into 'master'
soc/rtc: don't power down BIAS_I2C along with APLL if other PLL is used

See merge request !1568
2017-11-21 12:13:07 +08:00
Ivan Grokhotkov
d135a017f4 Merge branch 'bugfix/ut_startup_time' into 'master'
By default, don't print test menu in unit-test-app

See merge request !1565
2017-11-21 12:12:46 +08:00
Angus Gratton
dc994e9c88 ota example: Send a valid HTTP/1.0 request
Was sending an invalid HTTP/1.1 request which many web servers rejected.

Thanks @chegawara for pointing this out on IRC.

Ref:
https://github.com/espressif/esp-idf/issues/231#issuecomment-300287523
2017-11-21 09:02:58 +11:00
Angus Gratton
93c75b4978 Merge branch 'bugfix/update_BOD_voltages' into 'master'
esp32/Update brownout voltages

See merge request !1520
2017-11-21 05:37:55 +08:00
Yulong
729e0c2544 component/bt: Correct the SP_BLE_PWR_TYPE_CONN_HDL0-9 comment in the bt.h file to ESP_BLE_PWR_TYPE_CONN_HDL0-8 2017-11-20 08:05:35 -05:00
Ivan Grokhotkov
2eae440646 unit-test-app: reduce startup delay, clean up 2017-11-20 17:51:27 +08:00
He Yin Ling
81f806e676 test: change initial condition of unit test:
now UT App will not print test menu by default. it will print a prompt
to indicate it's ready. we can check this prompt instead of delay.
2017-11-20 17:51:27 +08:00
Ivan Grokhotkov
5117364c13 unit-test-app: don't print the list of tests by default
This reduces unit-test-app startup time when there are a lot of tests
included, speeding up unit tests in CI.
2017-11-20 17:51:27 +08:00
Ivan Grokhotkov
024e4c4337 Merge branch 'feature/openssl_cn_hostname_verification' into 'master'
openssl: Add CN hostname verification

See merge request !1554
2017-11-20 15:59:15 +08:00
Ivan Grokhotkov
18f4696371 Merge branch 'bugfix/pthread_test_mem_leaks' into 'master'
pthread: Fixes memory leaks and stack overflow in tests

See merge request !1563
2017-11-20 15:58:53 +08:00
Ivan Grokhotkov
4b8f55aadf Merge branch 'bugfix/minor_fixes_github' into 'master'
Minor fixes from github

See merge request !1555
2017-11-20 15:58:34 +08:00
Ivan Grokhotkov
3369f15fa3 soc/rtc: don't power down BIAS_I2C along with APLL if other PLL is used
If enable == false, and SOC_CLK_SEL == PLL, the code would would
erroneously set RTC_CNTL_BIAS_I2C_FORCE_PD. This change fixes the logic.
2017-11-20 15:27:16 +08:00
zhangyanjiao
9fbcc6f922 In STA mode, tcp_send cannot catch err after disconnect station 2017-11-20 15:19:19 +08:00
Angus Gratton
19598fa6d3 monitor: Fix Windows 10 crash since Fall Creators Update
Closes https://github.com/espressif/esp-idf/issues/1136
2017-11-20 17:47:43 +11:00
Andy Green
ae1f1e9b84 openssl wrapper: introduce X509_VERIFY_PARAM_set/clear_hostflags
This defines the OpenSSL X509_CHECK_FLAG_...s and the set/clear
accessors.  Since none of them are supported, the set / clear
accessor currently always does nothing and returns error.

This call is often part of the generic openssl user code to
set up certificate verification.  This patch allows it to
compile for ESP32 and decide at runtime what to do about
unsupported flags.

Merges https://github.com/espressif/esp-idf/pull/980
2017-11-20 16:25:57 +11:00
Angus Gratton
4395f7c94e Merge branch 'feature/spi_flash_lock_recursive' into 'master'
spi_flash: Expose recursive op_lock for atomic multi-part flash operations

See merge request !1556
2017-11-20 13:24:41 +08:00
Andy Green
effc6c6d0d openssl wrapper: introduce X509_VERIFY_PARAM_set1_host
This lets the user code set the mbedtls hostname using the standard OpenSSL
X509_VERIFY_PARAM_set1_host() API semantics.

The API takes an X509_VERIFY_PARAM pointer.  We use the fact that is
a composed member of the SSL struct to derive the SSL pointer.

The X509_VERIFY_PARAM_set1_host() is unusual in that it can accept a
NUL terminated C string as usual, or a nonterminated pointer + length.
This implementation converts the latter to the former if given, before
using it.

This is enough for user code to get the openssl wrapper to make
mbedtls confirm the CN on the peer cert belongs to the hostname used
to reach it, by doing, eg

	X509_VERIFY_PARAM_set1_host(SSL_get0_param(myssl), myhostname, 0);

Merges https://github.com/espressif/esp-idf/pull/980
2017-11-20 16:24:06 +11:00
Andy Green
0f02a38262 openssl wrapper: introduce SSL_get0_param
This adds the standard OpenSSL api to get a pointer to the SSL struct's
X509_VERIFY_PARAM.  We need this for the OpenSSL api to set the peer
hostname introduced in the next patch.

Part of https://github.com/espressif/esp-idf/pull/980
2017-11-20 16:23:18 +11:00
Angus Gratton
4b8c90bce0 doc: Unify the two sections about the "dialout" group on Linux
An addition to https://github.com/espressif/esp-idf/pull/1158
2017-11-20 16:10:38 +11:00
Angus Gratton
05dd46cd72 udp_perf/tcp_perf examples: Erase NVS if no more free pages
Follows same pattern as other examples.

Amendment to https://github.com/espressif/esp-idf/pull/1279
2017-11-20 16:07:52 +11:00
petermccloud
3b84c1e8fa fixed broken udp_perf example by adding nvs_flash
Merges https://github.com/espressif/esp-idf/pull/1279
2017-11-20 16:07:00 +11:00
petermccloud
0dea8bca46 fixed broken tcp_perf example by adding nvs_flash 2017-11-20 16:06:59 +11:00
Angus Gratton
f54e99008d spi: Add ESP_ERR_NO_MEM to spi_device_queue_trans() documentation
Ref https://github.com/espressif/esp-idf/pull/1243#issuecomment-343103414
2017-11-20 16:06:59 +11:00
Angus Gratton
7a924bd85a spi_flash: Expose an accessor the current SPI flash guard functions
Change places which uses g_flash_guard_default_ops to use this. Probably exact same data, but a bit
cleaner.
2017-11-20 15:54:31 +11:00
Angus Gratton
e72e5a9706 spi_flash: Make spi_flash op_lock/op_unlock recursive
Allows chaining of multiple flash operations as an atomic operation.
2017-11-20 15:54:31 +11:00
Darian Leung
91ccf588d1 esp32/Update brownout voltages
This commit updates the voltages of each brownout level in Kconfig.
Note that there is some variation of brownout voltage between each chip.
2017-11-20 12:26:13 +08:00
Ivan Grokhotkov
1fbd3ffac8 Merge branch 'feature/fatfs_r013a' into 'master'
fatfs: update to R0.13a

See merge request !1524
2017-11-20 10:16:39 +08:00
Ivan Grokhotkov
148130066f Merge branch 'feature/esp32_stack_protector' into 'master'
esp32: Adds Stack Smashing Protection Feature

See merge request !1537
2017-11-20 10:16:15 +08:00
Ivan Grokhotkov
9f7d632b74 Merge branch 'bugfix/spi_master_cs' into 'master'
fix(spi_master): fix the CS pin issue with GPIO 0.

See merge request !1536
2017-11-20 10:15:44 +08:00
Alexey Gerenkov
7ce945a9de pthread: Fixes memory leaks and stack overflow in tests
Also this commit replaces FreeRTOS list used for pthread internals
with simple one from rom/queue.h
2017-11-17 16:57:33 +03:00
Alexey Gerenkov
692a890232 esp32: Adds Stack Smashing Protection Feature
These changes add posibility to enable GCC stack protector via menuconfig
for all source files in project.
2017-11-17 12:08:36 +03:00
michael
02573a8dc9 fix(spi_master): release temporary memory when queue new trans timeout. 2017-11-17 12:15:19 +08:00
michael
13d38f10ee test(spi): internal connection test master with slave for address and command field. 2017-11-17 12:15:19 +08:00
michael
57e1d47ad7 fix(spi_master): fix the CS pin issue with GPIO 0.
Closes #909
2017-11-17 12:15:19 +08:00
michael
842292523f fix(spi_slave): add interrupt free into bus free process. 2017-11-17 12:14:40 +08:00
Raphael Luba
31711b5ac7 SPI: Document TIMEOUT return values
`spi_device_queue_trans` and `spi_device_get_trans_result` can return `ESP_ERR_TIMEOUT` – but this had not been documented.

Merges https://github.com/espressif/esp-idf/pull/1243
2017-11-17 10:50:54 +11:00
Dominik Palo
ad8ebe5b63 Update spp_client_demo.c
Merges https://github.com/espressif/esp-idf/pull/1229
2017-11-17 10:49:51 +11:00
Krzysztof Bociurko
4da90f09e7 freertos: minor issue in documentation snippet of queue.
taskYIELD was used in ISR context, but portYIELD_FROM_ISR should instead.

Merges https://github.com/espressif/esp-idf/pull/1187
2017-11-17 10:44:19 +11:00
Simon Werner
17b4973d47 docs: Improved documentation for /dev/ttyUSB0 issues
Improved documentation for /dev/ttyUSB0 issues.  This occurred in Ubuntu, but I'm sure is applies to most distributions.

Merges https://github.com/espressif/esp-idf/pull/1158
2017-11-17 10:43:02 +11:00
He Yin Ling
0c40b2ec9d test: collect performance for esp-timer / spinlock 2017-11-16 17:48:48 +08:00
He Yin Ling
879160de8c test: collect example https_request binary size 2017-11-16 17:48:30 +08:00